The Campaign of Pharaoh Shoshenq I in Palestine

Kevin A. Wilson
Professor of Biblical Studies
Lithuania Christian College
September 2004

A recent debate has been going on among our colleagues in the field of archaeology about the dating of 10th century B.C.E. strata at sites in Israel. The debate is over which strata should be assigned to the period of Solomon and which should be dated to the following period. On one side are those, such as Lawrence Stager and William Dever, who defend the traditional dating, which assigns the strata with excellent architectural remains to the Solomonic period, thereby confirming the biblical account of Solomon’s building activity. Others, such as Israel Finkelstein and David Ussishkin, date the layers with the monumental architecture to the 9th century, while viewing the earlier – and less rich – layers as those of Solomon.1 But no matter which side of the debate is taken, the same method is followed. Each relies heavily on identifying the destruction layers left by Shoshenq I – the biblical Shishak, who campaign in Palestine around 926 B.C.E. Once that layer is identified, the stratum to which it belongs is viewed as the Solomonic level since the Shoshenq campaign occurred just five years after the death of Solomon. Although the debate usually focuses on Megiddo, destruction levels at a large number of sites in Israel have been attributed to Shoshenq.2 This fact allows levels at multiple sites to be dated to the same period, and the Shoshenq campaign is therefore thought to provide an anchor for the archaeological sequence of the 10th century. Obviously, this is a greatly over-simplified statement of a complex issue, but it shows the importance of the campaign of Shoshenq in contemporary scholarship.

The campaign of Pharaoh Shoshenq I, the founder of the 22nd Dynasty who ruled from approximately 946-925 B.C.E., has been the focus of studies by quite a number of scholars this century. On the biblical side, scholars of such stature as Martin Noth, Benjamin Mazar, Sigfried Herrmann, and Gösta Ahlström have each produced articles on the campaign.3 And from the egyptological side, Kenneth Kitchen explored the campaign in his book on the Third Intermediate Period in Egypt.4 Although these scholars disagree on the details of the campaign, all take the same general approach to the source material. Each begins with a short study of the biblical accounts of the campaign, found in 1 Kings 14:25-28 and 2 Chronicles 12:1-12. After exhausting the small amount of information the Bible provides, they turn to the triumphal relief of Shoshenq found next to the Bubastite Portal at the Temple of Karnak in Luxor, Egypt.5 They use the topographical list that forms a part of that relief as a road map for the campaign. Each scholar arranges these toponyms – all of which are sites in Palestine – to reconstruct the line of march of the pharaoh’s army.

The reconstructions of the campaign offered by these scholars leave a couple of questions unanswered. First, while the biblical account of the campaign only mentions Jerusalem as the focus of the Egyptian attack, few sites in Judah are found in the triumphal relief. Instead, the list comprises sites in Israel and the Negev. Scholars explain this discrepancy by saying that 1 Kings is only concerned with Judah, so it ignores the part of the campaign that dealt with Egypt. This explanation is unsatisfying, however, since the book of Kings actually spends much more space discussing the reign of Jeroboam in Israel than it does describing the reign of Rehoboam in Judah. Why, then, would it leave out the portion of the campaign that took place in Israel? Second, these reconstructions do not explain why Shoshenq attacked Jeroboam, a man he had recently harbored as a political refugee. Some say Shoshenq made the campaign to punish Jeroboam for being a disobedient vassal.6 But although there is some information that suggests Jeroboam was a vassal of Egypt, there is no evidence that he ever rebelled.

Scholars’ use of the triumphal relief of Shoshenq suffers from several methodological problems as well. First, the triumphal relief comprises three parts: the relief scene itself, the accompanying inscription, and the topographical list. Yet those who have studied the Shoshenq campaign have mostly ignored the relief and the inscription, while focusing almost exclusively on the topographical list. This leads to problems in interpretation since such reliefs must be understood as a whole. Second, scholars have overlooked the fact that the triumphal relief of Shoshenq is actually one of many such triumphal reliefs.7 These reliefs were built by pharaohs throughout the New Kingdom, which means that the relief of Shoshenq is only one example of a genre that was used for many centuries. But without coming to a complete understanding of this genre in its many manifestations, it will be difficult to interpret any one example.

In order to rectify this situation, I undertook a study of several other triumphal reliefs. Twelve reliefs were selected for this study: six belong to Thutmose III, two were built by Seti I, two were left by Ramses II, and two belong to Ramses III. With the exception of the reliefs of Ramses III, which are found at Medinet Habu, all of the reliefs are at Karnak Temple, the same temple where the Shoshenq relief is located. Although a complete discussion of these reliefs is not possible in this forum, the following summary details the two most important conclusions of the survey.

First, it is clear that the topographical lists in these reliefs do not preserve the army’s route of march. This may be seen by comparing these topographical lists with known itineraries for Egyptian campaigns. The route of Thutmose III’s march, for instance, is known from his Annals, which are also inscribed on the walls of Karnak.8 The Annals give a prose account of his first campaign, which is the same campaign mentioned in the superscription to the topographical list in three of his triumphal reliefs. When the route of march from the Annals is compared with the topographical lists, it becomes apparent that the latter are not arranged according to the army’s itinerary. Gaza, one of the first cities of Canaan mentioned in the Annals, is not listed on the triumphal relief at all. The next two towns that Thutmose III passes are listed in two out of three of the topographical lists, but they fall far down the list.9 The order of these two names in the list is also different from that given in the Annals, with Yehem, which was reached first, coming after Aruna, which was reached later. Megiddo, which was visited after passing all of these towns, is the second town in the lists.10 And Kadesh, a town to which Thutmose III did not go, is first on the lists.11 In addition, the superscriptions to the topographical lists indicate that Thutmose III did not march to all these towns, but that they all assembled against him at Megiddo instead. The same results are also found when comparing the topographical lists of Seti I with known campaign itineraries. This is even more evident in the triumphal reliefs of Ramses III. His lists contains approximately 125 sites in Palestine, but not one of the cities where he is known to have campaigned in found in those lists.12 All these factors demonstrate that the topographical lists do not preserve the route of the pharaoh’s march.

Second, the survey of the triumphal reliefs indicates that these reliefs were not intended to preserve information about campaigns. This is seen when interpreting the relief as a whole and taking into account all three parts: the inscription, the relief scene, and the topographical lists. First, the inscriptions, which are set as the speech of Amun, do not mention particular campaigns, nor do they refer to individual battles. Instead, they make reference to general areas, such as Asia and Nubia, and to cultural groups, such as asiatics and bedouin. One of the phrases that occurs repeatedly is a reference to "all foreign lands," which is in contrast to battle reliefs, where the enemy and town are named specifically. The relief scenes have the same character. Each shows the pharaoh smiting a group of prisoners, but the scene depicted is not a battle relief. Nor are the prisoners closely identified. Usually, they are shown as a mixed group of Asiatics and Nubians, i.e. those enemies to the north and south of Egypt. They also hold stylized weapons that are symbolic of the enemies to the east and west of the Nile valley. The captives were probably meant to be understood as representing the hostile people that surrounded Egypt on all sides.13 This is, in fact, how they are identified in the inscription. A part of the text directly above the prisoners in most of the reliefs labels the scene as the smiting of the chiefs of all foreign lands.

If this is the interpretation of the relief scene and the inscription, then the topographical lists should be understood in this light as well. According to this interpretation, the topographical lists should be viewed as representing all foreign lands. So, while the inscription refers to the king as having subdued the entire world, the topographical lists provide a graphic illustration of this same thing. In other words, the inscription says that the pharaoh has conquered all foreign lands, the relief scene depicts the pharaoh simultaneously smiting captives from all foreign lands, and the topographical list lists cities in foreign lands that the pharaoh has defeated. Instead of being the record of a campaign, the triumphal reliefs show an idealized picture of the accomplishments of the pharaoh, who is portrayed as the conqueror of the known world.

These two results of the survey indicate that the triumphal relief of Shoshenq cannot be used as a source of information about the campaign mentioned in the Bible. His relief does not preserve the army’s itinerary, nor was it intended to commemorate the expedition. The Bible must therefore be viewed as the main source for the expedition. There is one other Egyptian text, however, that does shed a small amount of light on the campaign, namely a fragment from a stele of Shoshenq unearthed at Megiddo.14 This stele fragment is unfortunately not well preserved. With the exception of a few stock phrases glorifying the king, only the nomen and pre-nomen of Shoshenq are readable. This stele has often been taken as evidence that Shoshenq conquered Megiddo.15 The unspoken assumption behind that determination is the idea that the only reason a stele would be erected in a particular town would be to commemorate the pharaoh’s conquest of that town. While many stelae were set up on such occasions, stelae were also erected for other reasons. The second Beth-Shean stele of Seti I, for instance, records a campaign against the ‘Apiru and the Tayaru, who were attacking the town of Ruhama.16 But the stele commemorating this battle was set up in Beth-Shean, an Egyptian administrative center that was not directly involved in the fighting. The same is true for the Beth-Shean stele of Ramses II, which is a rhetorical stele that does not mention any specific battles.17 As far as is known from Egyptian sources, Ramses II never campaigned against Beth-Shean, and the fact that Beth-Shean had long been an Egyptian stronghold would mean that he had no need of campaigning there.

Yet although the stele of Shoshenq does not indicate that he conquered that town, it does show that he exerted some control over Megiddo. Unfortunately, it is not possible to state what type of relationship would have existed between Megiddo and Egypt since such stelae are found in towns over which the pharaoh exercised various types of control. In the Beth-Shean stelae, for example, it is obvious that the Egyptians had complete control over the city. It was used as an administrative center throughout the Late Bronze Age, as is attested in Egyptian texts and the archaeological record.18 At Kadesh, where another stele of Seti I is found, a different relationship is evident.19 Seti I had campaigned against the city, but after it was captured, it was not an Egyptian center. Instead, it was a vassal state within the Egyptian realm, which indicates slightly less direct Egyptian control over Kadesh than that which would have been exercised over Beth-Shean. The same situation is evident at Tyre, a vassal state where yet another Seti I stele was discovered. Tyre was never attacked by Egypt, however, and seems to have merely maintained its longstanding good relations with Egypt. This indicates that the stele of Shoshenq can be used as evidence for some form of recognition of Egyptian power by the Israelites at Megiddo, but it cannot be determined whether the pharaoh had conquered that city.

Having exhausted the Egyptian material, the biblical accounts must once again be viewed as the primary source for Shoshenq’s campaign. Unfortunately, these do not provide a great deal of information. The account in 1 Kings 14:25-28 only says that in the fifth year of Rehoboam, Shoshenq came up against Jerusalem and took the treasures of both the palace and the temple. As Noth pointed out long ago, the campaign itself is not even the focus of that passage. Its main concern is to explain why the gold shields of Solomon were no longer seen in the temple. The parallel version in 2 Chronicles 12:1-12 gives more information, including the reaction of the people in Jerusalem, but adds little of historical value.

In addition to these two passages, one other biblical text provides information about Shoshenq. This is the notice that Jeroboam took refuge in the court of Shoshenq after Solomon tried to kill him for treason. Jeroboam had been a high official in Solomon’s administration, but at some point he appears to have rebelled. When Solomon sought to put him to death, Jeroboam fled to Egypt, where he took refuge in the court of the pharaoh. After the death of Solomon, he returned to Israel, where he took part in the Shechem Assembly that rejected Rehoboam as king. Following this meeting, Jeroboam himself was made king.

With this new interpretation of the Egyptian material and the preceding summary of the biblical evidence, the following reconstruction of the campaign of Shoshenq and the events leading up to it can now be offered. The foreign policy of the 21st Dynasty in Egypt seems to have been rather mixed. On the one hand, the pharaohs appear to have had a political treaty with the United Monarchy of David and Solomon, as is evidenced by the marriage of Solomon to the pharaoh’s daughter mentioned several times in 1 Kings. At the same time, however, Egypt was not above harboring political refugees who were enemies of Israel, such as Hadad, of the Edomite royal family, who returned to Palestine to cause trouble upon the death of David. Military actions in Palestine may have also occurred. The Bible mentions, for instance, that Gezer was captured by the pharaoh, but later returned to Solomon as part of the wedding arrangement. It is likely, therefore, that Egypt followed a dual policy. Due to the internal weaknesses, Egypt was not in a position to openly oppose the United Monarchy, so the pharaohs ensured good relations between the countries through a political alliance. Yet they could not have been happy with such a strong military presence dominating their north-eastern border, so they worked behind the scenes to bring unrest and instability to Israel by backing political opponents.

With the passing of the 21st Dynasty and the beginning of the 22nd, this situation was still in place. This explains Shoshenq’s willingness to harbor Jeroboam, one of the chief opponents of the Davidic Monarchy. But with the death of Solomon, Shoshenq seems to have seized upon the opportunity to bring an end to the United Monarchy. When Jeroboam returned from Egypt to take part in the Assembly at Shechem, he probably had promises of support from Shoshenq. When Israel separated from Judah, it had the strength of Egypt behind it. This would explain the presence of the stele of Shoshenq at Megiddo since it could easily have been set up either to commemorate a treaty between the two nations or to signify the vassal status of Israel.

After the split of Israel and Judah, the Bible notes that there was constant warfare between Israel and Judah. It is in this context that the campaign of Shoshenq against Judah should be viewed. With the two states fighting, the pharaoh may have come to the aid of his ally, Jeroboam. Attacking only Jerusalem on this campaign, Shoshenq would have weakened Jerusalem, either through conquest or, more likely, through the looting of the temple and palace. In fact, the appearance of Shoshenq on the scene may lie behind the notice in 1 Kings 12:21-24 that says Rehoboam had planned to retake Israel by force, but quickly changed his mind, ostensibly at the urging of the prophet Shemiah. Having persuaded Rehoboam not to attack Israel, Shoshenq returned to Egypt, leaving Jerusalem with little will or resources to fight against Jeroboam.

From a foreign policy standpoint, this was a wise course of action for Egypt. Throughout the early 10th century, Egypt had not been strong enough to directly oppose the United Monarchy. When the chance to encouraged internal rebellion came along, however, Egypt jumped at the chance. The split of Israel and Judah left Egypt in a much better position. Before, the pharaohs found themselves with a strong military presence on their northern border, a situation which forced them into a relationship between equals with the strongest of these states. After the division of the United Monarch, the political landscape in Palestine changed. No longer was Egypt confronted by a strong united enemy. Instead, it now faced a collection of smaller states that were fighting among themselves. In addition, one of those states was somewhat dependant on Egypt. By supporting the rebellion of Jeroboam, Egypt had been able to change a defensive situation into a position of strength.

The beginning of this paper discussed the current debate over the dating of 10th century archaeological strata and the reliance on destruction levels left by Shoshenq. Although the current study does not decide the issue one way or the other, it does show that both sides are mistaken in searching for destruction layers left by the Egyptian campaign. As has been shown above, Shoshenq’s campaign was not as widespread as previously thought. Instead, it focused only on Jerusalem, and Jerusalem itself was not destroyed. This means that archaeologists will need to find another method for determining the date of 10th century strata. It also means that they will need to find other suspects for the destruction layers previously assigned to Shoshenq. This should not be difficult, however, since the 10th and early 9th centuries saw a great deal of fighting in Palestine, both between Judah and Israel, as well as between Israel and neighboring states. And, of course, the Israelites kingship changed hands several times through military coups during that period, so the causes of these destruction may have been internal. In any event, it is clear that the campaign of Shoshenq can play little role in the dating of 10th century archaeological strata.

Book form of Campaign of Pharaoh Shoshenq I into Palestine will be published by J C B Mohr (Paul Siebeck) in 2005

Footnotes

(back)1 Israel Finkelstein and David Ussishkin, “Back to Megiddo,” Biblical Archaeology Review 20.1 (January/February 1994): 26-43; David Ussishkin, “Notes on Megiddo, Gezer, Ashdod, and Tel Batash in the Tenth to Ninth Centuries B.C.,”  Bulletin of the American Schools of Oriental Research 277/278 (1990): 71-91.

(back)2 Amihai Mazar lists ten destructions attributed to Shoshenq: Timnah, Gezer, Tell el-Mazar, Tell el-Hama, Tell el-Sa‘idiyeh, Megiddo, Tell Abu Hawam, Tel Mevorakh, Tell Michal, and Tell Qasile.  Amihai Mazar, Archaeology of the Land of the Bible:10,000-586 B.C.E. (New York: Doubleday, 1990): 398.

(back)3 Martin Noth, “Die Wege der Pharaonenheere in Palästina und Syrien IV,”  Zeitschrift des Deutschen Palästine-Vereins 61 (1938): 277-304; Benjamin Mazar, “Pharaoh Shishak’s Campaign to the Land of Israel,” Vetus Testamentum Supplement 4 (1957): 57-66; reprinted in Benjamin Mazar, The Early Biblical Period: Historical Studies, ed. Smuel Aituv and Baruch A. Levine (Jerusalem: Israel Exploration Society, 1986): 139-150; Siegfried Herrmann, “Operationen Pharao Schoschenks I. im östlichen Ephraim,” Zeitschrift des Deutschen Palästine-Vereins 80 (1964): 55-79; Gösta W. Ahlström, “Pharaoh Shoshenq’s Campaign to Palestine,” History and Traditions of Early Israel: Studies Presented to Eduard Nielsen, ed. André Lemaire and Benedikt Otzen (Lieden: E.J. Brill, 1993): 1-16; see also Frank Clancy, “Shishak/Shoshenq’s Travels,”  Journal for the Study of the Old Testament 86 (1999): 3-23.

(back)4 Kenneth A. Kitchen,  The Third Intermediate Period in Egypt (1100-650 B.C.),  2nd ed. with supp. (Warminster: Aris and Phillips, 1986).  The main section is found on pp. 292-302, while  an excursus surveying previous works and discussing the identification of name rings in the topographical list is found on pp.432-47.

(back)5 The Epigraphic Survey, Reliefs and Inscriptions at Karnak III: The Bubastite Portal, Oriental Institute Publications, vol. 74 (Chicago: University of Chicago Press, 1954).

(back)6 Mazar 147 (Mazar refers to Rehoboam in this section, though it is clear from the context that “Rehoboam” is a typo for “Jeroboam.”); Kitchen 298; Ahlström 14.

(back)7 A triumphal relief as defined here must include all three elements: smiting scene, inscription, and topographical list.  These elements are not confined to triumphal reliefs, of course.  The smiting scene is known from as far back as the Pre-Dynastic Period and continued in use through Roman times.  For a survey of the smiting scene throughout Egypt’s history, see Emma Swan Hall, The Pharaoh Smites his Enemies: A Comparative Study, Münchner Ägyptologische Studien 44 (Berlin: Deutscher Kunstverlag, 1986).  Topographical lists are found in a variety of locations, including temple reliefs and statue bases.
      In the Theban area alone, at least eleven additional sets of triumphal reliefs are known.  These include:

•       Thutmose III: west face of the 6th Pylon at Karnak.  K C 56 and K C 152 , in Harold Hayden Nelson,  Key Plans Showing Locations of Theban Temple Decorations, Oriental Institute Publications 56 (Chicago: University of Chicago Press, 1941).

•      Thutmose III: south face of the 7th Pylon at Karnak. K G 88 and K G 90 in Nelson, Key Plans.

•      Thutmose III: north face of 7th Pylon at Karnak.  K G 40 and K G 43 in Nelson, Key Plans.

•      Amenhotep II: south face of the 8th Pylon at Karnak.  K G 143 (east) and K G 145 (west) in Nelson, Key Plans.

•      Seti I: exterior of the northern wall of the Hypostyle Hall at Karnak.  K H 6 (west) and K H 14 (east) in Nelson,  Key Plans.

•      Ramses II: exterior of the western wall of the court of Ramses II at Luxor.  Nelson Key Plans, L G 44 (northern) and LG 49 (southern).

•      Ramses II: exterior of the southern wall of the Hypostyle Hall at Karnak.  K O 44 (west) and K O 50 (east) in Nelson, Key Plans.

•      Ramses III: north face of his station temple in the forecourt of Karnak.  Nelson Key Plans, K K 10 (east) and K K 1 (west).

•      Ramses III: exterior face of the 1st Pylon at Medinet Habu.  MH A 39 (South) and MH A 31 (North) in Nelson, Key Plans.

•      Shabaka: west face of the Ethiopian Pylon of the smaller temple at Medinet Habu.  Nelson Key Plans, MH B 223 (north) and MH B 232 (south).

•      Nectanebo I (usurped by him from an earlier Saite pharaoh): east face of the Saite Portico on the smaller temple at Medinet Habu.  Nelson Key Plans, MH B 267 (north) and MH B 270 (south).

(back)8 The text of the Annals is transcribed in Urk. IV, 645-67.  A convenient translation of the Megiddo section is available in Miriam Lichtheim, Ancient Egyptian Literature, vol. 2 (Berkeley: University of California Press, 1976): 29-35.

(back)9 Yehem is no.68 in Lists A and B, while Aruna is no.27 on Lists A and C.

(back)10 Megiddo is not found in List C, but the first three names of that list are not preserved.

(back)11 Donald Redford reached similar conclusions that he related in passing in a work devoted to Egyptian day-books.  He commented that “the progression of sites, when plotted on a map, produces such a curiously meandering line at times that one might easily be led to the further supposition that the field commander of the Egyptian army was drunk.”  Donald B. Redford, Pharaonic King-Lists, Annals, and Day-Books: A Contribution to the Study of the Egyptian Sense of History, SSEA Publication IV (Mississauga, Ontario: Benben Publications, 1986): 125.

(back)12 Ramses III campaigned in the following areas: Amor (amr; KRI V, 40:1), Zahi (Dh; KRI V, 30:5 and 40:7), Tunip (tnp; KRI V, 78:15), Ereth (irT; KRI V, 79:12), and Naharain (nhrn; KRI V, 88:8).  He also fought several groups of “Sea Peoples” on these campaigns, including the Peleset (prST; KRI V, 40:3), the Tjeker (Tkr; KRI V, 40:3), the Sheklesh (SkrS; KRI V, 40:3), the Denyen (dini; KRI V, 40:3-4), and the Weshesh (wSS; KRI V, 40:4).  None of these groups appear in the topographical lists either.

(back)13 On this idea, see G. Belova, “The Egyptian’s Idea of Hostile Encirclement,”    Proceedings of the Seventh International Congress of Egyptologists, ed. C.J. Eyre, Orientalia Lovaniensia Analecta 82 (Leuven: Uitgeverij Peeters, 1998): 143-148.

(back)14 C. S. Fischer, The Excavation of Armageddon, Oriental Institute Communications 4 (Chicago: University of Chicago, 1929): 12-16 and Fig. 7b.;  R. S. Lamon and G. M. Shipton, Megiddo I.  Seasons of 1935-39, Strata I-V, Oriental Institute Publications 42 (Chicago: University of Chicago, 1939): 61 and Fig. 70.

(back)15 Noth 281; Mazar 141; Kitchen TIP, 299; Ussishkin 74.

(back)16 Kenneth A. Kitchen, ed.,  Ramesside Inscriptions: Historical and Biographical, vol. 1 (Oxford: B. H. Blackwell, Ltd., 1975): ?  A translation may be found in Kenneth A. Kitchen, Ramesside Inscriptions: Translated and Annotated, vol. 1 (Oxford: Blackwell Publishers, 1993): 12-13.

(back)17 Kenneth A Kitchen, Ramesside Inscriptions: Historical and Biographical, vol. 2 (Oxford: B. H. Blackwell, Ltd., 1979): 150-151.  The text is translated in Kenneth A. Kitchen, Ramesside Inscriptions: Translated and Annotated, vol. 2 (Oxford: Blackwell Publishers, 1996): 27-29.

(back)18 See Frances W. James, Patrick E. McGovern, et al., The Late Bronze Age Garrison at Beth Shan: A Study of Levels VII and VIII, 2 vols.  (Philadelphia: University Museum of the University of Pennsylvania, 1993).

(back)19 Kitchen, Inscriptions, ?  Translated in Kitchen Inscriptions: Translated, 20.  Unfortunately, very little of the stele remains, and just enough text is preserved to show the nomen and pre-nomen of Seti I.